Sunday, January 15, 2006

Genetics of Sugarcane Quality Traits and Breeding for Quality Improvement


The quality traits in sugarcane depend on the end use of the cane in the industry rather than a common list of quality traits.  The main parameters are listed in the table below.

Major quality traits in sugarcane
Quality traits
Sugar Industry
Sugar recovery
Sucrose content in juice
Purity coefficient of juice
Higher juice volume
Higher total sugars
Paper Industry
Higher cellulose
Lower Ash content
Lower Lignin content
Lower Pith content or higher Fibre/pith ratio
Lower colouring matter in bagasse
Higher L/D Ratio of ultimate fibres

The novel idea of using the wild species Saccharum spontaneum (as male with S. officinarum as female) in sugarcane breeding with the objectives of incorporating gene complexes for biotic and abiotic stresses and for high biomass production resulted in a tremendous level of useful genetic variability including that of better cane quality, sugar content and incorporation of rationing ability. This paved the way for revolutionising sugarcane agriculture and sugar industry not only in India, but also in most of the sugarcane growing countries of the world. Most, if not all, of the present day varieties world-over have the Coimbatore bred canes in their genealogy.

In India, Co 419 released in 1933 became the most popular variety in tropical India and was rightly hailed as the 'wonder cane' the world-over. Co 997 and Co 1148, released in 1950s, became the ruling varieties in Andhra Pradesh and North India respectively. Co 1148 remained the most predominant variety for over four decades in sub-tropical belt and was hailed the 'Wonder Cane of North India'. Co 62175 became the most sought after variety by jaggery farmers owing to its heavy yield. Varietal evaluation conducted across seasons and different months of juice analysis helped in the identification of varieties with high sucrose levels such as Co 7204, Co 7704, CoA 7601, CoC 671, Co 8336, Co 8338 etc.

In the USA, evaluation of F1 crosses between commercial hybrids and Miscanthus and Erianthus (as male parents) indicated that thin stalk and low sucrose content were strongly influenced by Miscanthus and Erianthus. However, mean sucrose content and purity of F2 and BC1 seedlings of crosses between commercial hybrids and Miscanthus spp., Miscanthidium spp. and Erianthus spp. were markedly improved over that of F1 hybrids, but mean stalk diameter was still very small.

In Taiwan, sugar content in a Saccharum × Miscanthus hybrid showed a steady increase in successive nobilized generations. Clones selected from F1 Saccharum × Miscanthus hybrids were productive in fibre yields, with better pulp sheet as compared to a commercial control. Some clones from BC1 crosses involving sugarcane and sorghum had sucrose content equivalent to commercial controls and proved to be early maturing (TSRI, 1995-96).

In sugarcane breeding, the available evidences indicates that selection of parents based on their juice quality would be successful in increasing the juice quality of the succeeding generation, but selection of parents followed by progeny testing would seem to be necessary for increasing cane yield. Studies on milling characters showed that starch content in juice, percentage of reducing sugars, fibre percent fresh weight has high heritabilities and hence both parental selection and individual selection would be effective for these characters. Roach (1968) found highly significant male × female interaction effects for yield and sucrose in S. officinarum × S. spontaneum crosses. Heterosis was observed for early growth, stalk height, cane yield, flowering, pollen production and level of reducing sugars.

Inbreeding depression for components of cane yield viz., stalk diameter, stalks per plant, and stalk length was reported but there was no apparent depression for brix. This suggested that most of the genetic variance for brix be additive in nature.  However, it is important that non-additive genetic variance for cane yield and its components is required in sugarcane. This indication is supported by the few studies that have provided estimates of additive and non-additive genetic variances (Hogarth, 1987).

Studies on 14 crosses between commercial hybrids (female) and S. spontaneum (male) and found families to differ significantly for stalk number, stalk diameter and cane quality characters. Narrow-­sense heritability estimated from the partitioning of variances between and within family ranged from relatively low for dry matter % cane (Brix % cane + fibre % cane) (0.17) to moderately low for stalk number and fibre % cane (0.39 and 0.38, respectively) and moderately high for cane quality characters (0.67-0.90) and stalk diameter (0.75). These estimates were generally higher than those obtained from commercial × commercial crosses. The distribution of Brix, sucrose content, juice purity and fibre content in crosses between commercial hybrids and S. spontaneum clones were continuous and controlled by polygenes, and additive genetic variance was more important than dominance genetic variance for all four characters.

In experiments using diallel crosses, specific combining ability (indicating non-additive genetic variance) has generally been greater than general combining ability for both cane yield and sugar content as measured by brix. However, due to the possibility of self-pollination in such crosses, these results should be interpreted cautiously.

Estimates of heritability and clonal repeatability on an individual basis for three milling characters were reported in sugarcane.

Heritability and clonal repeatability of milling characters in sugarcane
Clonal Repeatability
Starch (log ppm)
0.53 +/- 0.14
Reducing sugars % extract
0.31 +/- 0.13
Fibre % fresh weight
0.34 +/- 0.15

This indicated that progress from selection of parents should have some value, particularly for starch content. Progress from clonal selection should be very good for all characters even on an individual basis. Studies on the inheritance of ash in juice showed that for ash percent in juice, 95% of the genetic variance was additive. Thus, there should be no difficulty in breeding for this character, if it is deemed desirable. However, the authors pointed out that selection based on ash content might reject many varieties with good agronomic characters.

Increasing fibre content of commercial varieties but still maintaining an acceptable sugar content was thought to be an attractive economic option that could cut down on imported coal and also reduce CO2 emissions. In the case of fibre percent fresh weight. 80% of the genetic variance is additive, and the character is moderately heritable on an individual basis (h,2 = 0.45), indicating that selection of parents for low or high fibre content should be effective in producing the progeny with desirable fibre content.

Evaluation of S. spontaneum, S. robustum, S. sinense, S. officinarum, commercial hybrids and Erianthus spp. for cane quality components, Brix % cane (total solids of juice), pol % cane (apparent sucrose), fibre % cane and dry matter % cane, by the method described by Saint Antoine and Froberville (1964). A ratio of pol : fibre (PF) was derived from the pol % dry matter and fibre % dry matter. Juice purity was also derived from the ratio of Brix % cane and pol % cane. These studies revealed high levels of fibre and dry matter content in S. spontaneum clones as compared with S. officinarum. Commercial hybrids ranked very close to S. officinarum, and the Erianthus group ranked very close to S. robustum. Brix % cane and pol % cane were the highest in S. officinarum and commercial hybrids whereas sucrose content of the Erianthus group was even lower than that of S. spontaneum. The PF ratio showed equal partitioning of dry matter between pol and fibre in the S. officinarum and commercial clones.

Many correlations between agronomic and milling characters were found in sugarcane. The genotypic correlation between cane yield and fibre percent was 0.57 and for sucrose percent and fibre was –0.76. Significant positive correlations between fibre content and brix was also reported. A very weak correlation between fibre content and brix in juice as well as negative correlations between brix and ash were also reported in sugarcane. This indicated that selection for brix might produce varieties with lower ash content.

With further developments, microsatellite markers could soon become a tool for genetic fingerprinting, genetic mapping and to assist in selecting specific, genetically diverse parents for use in introgression. The development of marker assisted selection would have a major potential use in introgressing a single quantitative trait locus (QTL) or multiple QTL from an ancestral species into commercial cane. The search for molecular markers for sucrose genes to assist in breeding and selection could reduce the number of backcrosses necessary to recover the recipient genotype near to the donor target genes.